Estimation of Accidental Coincidences in PET

Accidental coincidences are one of the main sources of image degradation in positron emission tomography. It is possible to compensate for their degradation effects, but an accurate method to estimate the randoms rate is required. Two conventional methods are used for random rate estimation: the "singles rate" method and the "delayed window" method. In this work we propose a mathematical model that describes the process of accidental coincidence formation. By using it, we are able to predict the correct number of randoms as well as the estimations provided the singles rate and delayed window methods. The model is used to propose a novel estimation method: the "singles-prompt". The aim of this work is to assess the performance of the singles-prompt method and, specially, the model capabilities at several levels. The results agree with all the predictions of the model. In particular, the singles rate and delayed window estimations behave as described by the model and the singles-prompt method estimates the correct number of randoms regardless of the source distribution and total activities.